Forced decision results of MSTAR dataset.
In this chapter, we present a cognitive radar architecture based on the three-layer model by Rasmussen. The skill-based-layer is characterized by adaptive signal-processing approaches and target matched waveforms. The rule-based-layer comprises reactive execution of optimal illumination policies and resource-management. The knowledge-based layer allows for long term, goal-oriented mission- and trajectory planning. Each layer is illustrated by example algorithms and applications for implementation.
- adaptive filters
- cognitive systems
- closed-loop controllers
- signal processing
- system architectures
Modern multifunctional radars with electronic beam-steering (AESA) provide many degrees of freedom to point the antenna beam, usage of the electromagnetic spectrum and waveform selection (Figure 1). Complex surveillance and reconnaissance scenarios require increased automation and suited man-machine-interfaces, which is enabled by the cognitive radar approach [1, 2, 3].
In this article we explain a cognitive radar architecture developed at the Fraunhofer FHR based on the three-layer-model of Rasmussen . In the following, we will first introduce the concept of cognitive automation and derive our cognitive radar architecture. For each cognitive subfunction several technologies for realization are discussed and illustrated by example applications.
2. Cognitive automation for radar
The concept of Dual-Mode Cognitive Automation  is well suited to deal with the challenges of highly automated radar systems. As shown in Figure 2, intelligent software-agents (depicted as robot-heads) can be introduced into the work equipment to increase the level of automation under the supervisory control paradigm  .
Alternatively the software-agent can cooperate with the human operator in the sense of an intelligent assistant system . Even though the cognitive radar architecture can be used for both approaches, we will focus on the more traditional supervisory control role in the following.
3. Three layer model of a cognitive radar architecture
The three-layer model of human cognitive performance published by Jens Rasmussen in 1983 is widely used in human factors , cognitive psychology and robotics [9, 10]. As shown in Figure 3 the complex process of human cognition is simplified and broken down into cognitive subfunctions (shown as gray boxes) with the indicated flow of information. The Rasmussen-model distinguishes three layers of cognitive performance with increasing level of abstraction.
For the development of a cognitive radar architecture in analogy to the Rasmussen-model, each cognitive subfunction had to be mapped into five different radar-technologies as shown in Figure 3.
Modern radar system can generate arbitrary waveforms in real-time. This allows for transmit signals to be matched to the target transfer function or the electromagnetic spectrum as explained in Sections 4.1 and 4.2. Perception tasks of a radar comprise signal-processing and classification aspects. We use a machine learning approach that is illustrated in Section 4.3. Rule-based behavior in a radar is emulated by using optimal control policies or resource management approaches as shown in Section 5.1 or Section 5.2. Knowledge-based behavior can be implemented using Bayesian networks or automated planning algorithms. We show an example for robot-trajectory planning in Section 6.1.
The skill-based layer represents the basic signal-generation and processing capabilities of the radar system. It operates on the smallest timescale in the architecture in a continuous processing loop. Below, we give an example for adapting the transmit waveform to the target-transfer function using arbitrary waveform generation capabilities. As an extension, the waveform can further be adapted to the electromagnetic spectrum that has to be continuously sensed.
4.1. Matched illumination
If a priori information about a target is available, it is possible to optimize the transmission waveform for this target. Advantages arise for example by discriminating two classes of targets or by reducing resources of the sensor. One example is the reduction of the required bandwidth, if the available a priori information about the target is comprehensive.
In order to resolve the size of the object, two transmission frequencies are sufficient to estimate the extension of two scattering points with a spacing of
realize. In a cognitive radar system, the a priori knowledge of the target can be presupposed by previous measurements and is assumed to be predicted for the next time step. An adapted waveform can be used to update the target track with respect to its extension by a lower allocation of the bandwidth.
Assuming a linear, time invariant channel with additive white Gaussian noise
If the target characteristic
with the constraint of an energy limited transmission signal and the Hermitian correlation matrix . One possibility to solve this optimisation problem is the Lagrangian multiplier method
Eq. (4) is obviously an eigenvalue equation where the Lagrange multiplier
To gain a better understanding of the solution, the basic example of Figure 4 is presented for the two dominant scattering points
The corresponding frequency spectrum to Eq. (5) is
The target impulse response fluctuates due to the interference of the scattering centers with a period of and depends therefore on the target dimension as already visualized in Figure 4. The phase shift of the second point target causes a shift of all frequency maxima (see also Figure 5(a)). Processing an eigenvalue decomposition according to Eq. (4) to obtain the optimal waveform for this example (see Figure 5(b)).
Comparing this basic results with the solution of the eigenvalue decomposition, it is obvious that both frequency spectra are related to each other. If all frequency components of the target impulse response have comparable magnitudes, the frequency characteristic of the largest eigenvalue is similar to the target frequency spectrum and the target extension respectively. According to Eq. (6) the period of the target extension corresponds to a frequency of
MHZ. The envelope of the transmission wave is related to the phase difference between
In order to distinguish between targets an adapted waveform can be used to improve the discrimination between two types of target classes . A binary hypothesis test is one method to discriminate between target classes by evaluating the received signal
The energy is focused in the spectral area where the both target deviations are predominant.
with the variance of the complex noise. The deflection for the likelihood ratio test defines the effective difference of the likelihood centers and represents the output signal to noise ratio
It is possible to achieve the same performance of the receiver operating characteristic (ROC) curve, with a different test statistic of that likelihood ratio but different deflections . That is why a higher deflection is related to a better discrimination and a lower sensitivity with respect to an suboptimal threshold. Figure 6 shows the results of the binary test for a linear frequency modulation (LFM) and the optimized waveform for two Gaussian targets with the same extension and distance. The deflection between both classes increases for the optimised waveform leading to a lower intersection are of the test statistic for the hypothesis and the alternative. This facilitates a better separability as well as a lower false alarm rate for the same detection probability.
One example of adapting the waveform to the environment is the support of the classification and saving resources like the bandwidth. Applications like interference mitigation can also be executed in the skill-based layer by combining spectrum sensing algorithms with matched illumination.
4.2. Spectrum sensing
Due to the fact that wireless communication technologies are of significant importance in modern times, the available radio frequency spectrum has become a valuable resource for radar. For example the U.S. department of commerce  has decided to allocate parts of the S-band (1695–1710 MHZ and 3550–3650 MHZ) to wireless communication. Another example are parts of the C-band (5150–5350 MHZ and 5470–5725 MHZ) which are used by weather radars but are also used by 5GHz-WiFi  now. On the other hand these bands, although allocated, are underutilized providing opportunities for secondary (unlicensed) users to share the bands without harming the primary users. The other way round, a similar problem arises when the radar suffers interference from other users or even active jamming. Especially the first is a problem for ultra wideband radars like ground penetrating radars which naturally operate in partially occupied frequency bands. In the future these problems will become even worse and hence future cognitive radar systems must be able to operate in spectrally dense environments. Spectrum sensing techniques from cognitive radio provide algorithms to identify spectrum opportunities, i.e. to decide if a frequency band is occupied or not. With this information a cognitive radar can adapt dynamically its bandwidth, frequency and other transmit parameters to the radio frequency environment.
A significant number of studies dealing with spectrum sensing algorithms exists and hence we only give a brief overview here. For a comprehensive overview the reader is referred, for example, to the surveys [17, 18]. Spectrum sensing algorithms can be split into wideband and narrowband algorithms. Almost all narrowband spectrum sensing methods are statistical hypothesis tests usually written as
Most modern modulations like OFDM or QAM have cyclostationary properties. For details on a test statistic see . These methods offer high detection probabilities even in low SNR regions and are blind in the sense that they do not need information about . The price is a very high computational complexity and prior information about the used modulation. Completely blind methods, i.e. absolutely no prior information is necessary, are based usually on a multi antenna system. The data from the different channels is used to estimate a covariance matrix and from its characteristics e.g. eigenvalues a test statistic is build, see .
Because the channel state may change between the sensing and transmitting a prediction step after the sensing is helpful or even needed. For this purpose hidden Markov models are used in Ref.  and additionally multilayer perceptrons and recurrent neural networks are considered in Ref. . Especially the neural networks perform well in simulations with a prediction accuracy of about 0.8 to 0.9.
In contrary to the narrowband band spectrum sensing the wideband spectrum sensing methods divide a band into occupied and unoccupied subbands. The most obvious method for classifying a wideband is to split it into fixed subbands (using a FFT or sweep and tune) and perform narrowband sensing in each one. But there are also native wideband spectrum sensing methods like a wavelet based approach, see .
If the radar is the primary user and avoiding or reducing interference is the only goal of the spectrum sensing, it is not necessary to decide if a channel is occupied or not. It is sufficient to use the channel with the least interference. But if a lot of interference is present, a compromise between bandwidth (resolution) and interference must be made which leads to an optimisation problem, see Refs. [26, 27].
After each sensing period, a suitable and adaptable waveform must be generated taking the information from the sensing step into account, essentially bandwidth and center frequency. For example, this can be multiple or notched chirps filling the unoccupied bands or a stepped FM waveform which avoids the occupied frequencies, see . A combination with the matched illumination approach presented in Section 4.1 can be considered, too.
Building an experimental radar system with spectrum sensing capabilities is a challenging task. The computational complexity of some algorithms can be a burden and the additional sensing time, i.e. gathering the samples and computation time must be taken into account, causing a reduced duty cycle or pulse repetition frequency. In Ref.  a radar system employing spectrum sensing and matched illumination was implemented using an Ettus USRP X310 software defined radio. In a test environment about 10 dB noise floor reduction were achieved using spectrum sensing and a notched chirp.
4.3. Classification with deep learning techniques
The transition from the continuous stream of incoming row-data towards a symbolic representation of objects, which forms the basis for higher-level cognitive processing, is typically achieved using pattern recognition or classification techniques. As shown in Figure 3, machine learning approaches comprise subsymbolic feature formation processes that separate characteristic signal features in a higher-dimensional space. In this feature space, it is easier to recognize certain target classes to create an abstracted situational picture within the cognitive radar system.
4.3.1. Convolutional neural networks
Convolutional Neural Networks (CNNs) are inspired by the visual system of the brain and are part of the deep learning research field. For many years, CNNs were the only type of deep neural network that could efficiently be trained due to their structure using the technique of weight sharing . The basic structure of the network used in the presented architecture is shown in Figure 7.
CNN’s are a special form of multi-layer perceptrons, which are designed specifically to recognize two-dimensional shapes with a high degree of invariance to translation, scaling, skewing, and other forms of distortion . This invariance is achieved by an alternation of convolutional and subsampling layers, in which the neurons are organized in so called feature maps. All neurons in each of these feature maps use the same weights and are connected to a local receptive field in the previous layer. With this weight sharing technique, the number of free parameters is dramatically reduced compared to a fully connected network, what should lead to a better generalization of the network.
In the first convolutional layer, each neuron takes its inputs from a local receptive field in the input image and the output values of each feature map, which are visible in Figure 7, represent the intensity of one specific local spatial feature. The features, i.e. the weights of the neurons, are learned during the training process and since the receptive fields of neighboring neurons in the feature maps are shifted only by one pixel in the corresponding direction in the input image, the output values of each feature map correspond to the result of a two-dimensional correlation of the input image with the learned weights of each particular feature map.
In the input image of Figure 7 one target is visible in the center of the image. The correlation with the different kernels is visualized for three examples. The learned kernels are depicted inside the black squares on the input image and the result of the correlation can be seen in the feature maps of the first layer.
The second layer of the network is a subsampling layer and performs a reduction of the dimension by a factor of four. With this reduction the exact position of the feature becomes less important and it reduces the sensitivity to other forms of distortion . The subsampling is done by averaging an area of 4 × 4 pixels, multiplying it with a weight
The third layer is a convolutional layer again and relates the features found in the image to each other. This layer is trained to find pattern of features, which can be separated by the subsequent layers and discriminate the different classes. The output of this layer is the internal representation and can be considered as feature vector found by the network for the given input image.
The last two layers of the network form the decision part of the system and are fully connected layers, which use the output values of the third layer as features for classification. The last layer consists of as many neurons as classes have to be separated, in our case ten. The classification is done by assigning the corresponding class of the neuron with the highest output value.
One cost function for neural networks trained with the back propagation algorithm is the mean square error (MSE) of the training set. The MSE is the mean value of the quadratic loss function
The MSE is also called the empirical risk with respect to quadratic loss and classifiers using this error as a performance measure are said to implement the empirical risk minimization (ERM) .
with the constant
Further details of the algorithm and the approximations that are done to compute the derivatives can be found in Ref. .
4.3.2. Regularizations and adaptive learning rates
One feature of the presented network is the use of
which can also be written as
The use of momentum should have a positive effect on the behavior of the training algorithm and may prevent the algorithm from converging to a local minimum of the error function . Another important regularization method used in this network is the max-norm regularization of the weights of the network. For this regularization the Frobenius norm of each kernel in layer one and three is calculated after the weight change at every iteration and if the norm is larger than a certain value
So far the learning rate in (16) is only determined by the characteristics of the data itself and the error it produces at the output of the network. Another important factor could be meta-information available about the training set. We give here an example of a priority class, which means that we have one target in our database that should always be classified correctly with the additional cost that we might produce more errors in other classes. To incorporate these priority classes into our network, the representation in (22) is used. The general idea is to increase the learning rate
If this term is included in the formula for the weight change Δ
The need for a different weighting of classes is also discussed in Ref. , where it is mentioned that the different costs of misclassification should be part of the classifier design. The way we used here to include this prior knowledge into our target recognition system was also mentioned in Ref.  for Support Vector Machines, where the idea was to penalize the samples of less represented classes higher than others.
To show the benefit of this adaptive learning strategy we show an example of the ten class moving and stationary target acquisition and recognition (MSTAR) data  in Figure 8. In this example the learning rate of class four is multiplied with different weightings between one, which means no priority, and ten.
Without any weighting, this class has compared to the other classes a rather low correct classification rate calculated with respect to the number of input images Pccin (curve with round markers). This value gives the amount of input images that belong to class four and are actually classified as class four. The curve with the square markers in the plot gives the probability of correct classification with respect to the number of output images Pccout, which gives the amount of images that are classified as class four really belong to class four and is thus an indicator on the reliability of the classification. Summarized over all classes, both indicators lead to the same result, the correct classification rate Pcc of the curve with the triangular markers. From the plot can be seen that Pccin shows a steep increase at small values of
4.3.3. Combination of convolutional neural networks with support vector machines
An often mentioned benefit of Support Vector Machines (SVMs) is the high generalization capability in comparison to neural networks. The high generalization of SVMs is achieved by a training strategy called
To use the high generalization of SVMs in our classification framework, we replace the last two layers of the CNN in Figure 7 with SVMs. In this way we can use the convolutional feature extraction with the invariance to different forms of distortion and a classifier with high generalization. As input for the SVMs, the output values of the third layer are used. The final structure of the classifier is shown in Figure 9.
A SVM can only separate between two classes, for this reason the training set must be split for each SVM into two parts, one part containing the class that should give a positive result at the output of the SVM and one part containing the remaining training set that should give a negative result at the output. SVMs trained in that way are working in the one vs. all classification scheme, which means that as many SVMs as classes that need to be separated are necessary. For the actual classification of a SVM, a kernel is used to transform the data to a high dimensional space in which it is more likely that the problem can be linearly separated. Two common kernels are polynomial (including linear and quadratic kernels) and radial basis functions (RBFs). In Table 1 a small example of the MSTAR database is shown and it can be seen that the already very high correct classification rate of the CNN can be further increased with the use of SVMs as classifier.
|CNN feature extraction and polynomial SVM||98.19%||1.81%|
|CNN feature extraction and RBF SVM||98.28%||1.72%|
The results shown here are so called
Based on the abstracted situational picture derived by signal-processing and machine learning techniques, the cognitive radar system has to react to the perceived scene. Below, we illustrate a MDP based scheme to execute a-priori known, optimal illumination policies. In multifunctional radars, a radar resource manager has to schedule the individual illuminations into a serial radar timeline.
5.1. Optimal illumination policy
Markov-decision-processes (MDPs) are widely used in robotics to derive optimal control policies in stochastic environments. An agent in state
Three classes of targets
The resulting multi-stage illumination policy is shown in Figure 12. Initially the target is illuminated with mode 2 and classified. Depending on the result
5.2. Radar resource management
The illumination-strategy in Figure 12 requires up to five consecutive illuminations of a target. As indicated in Figure 1, a multifunctional radar must simultaneously carry out additional tasks, in particular search for the new targets and track known targets. Since a shared aperture is used, the radar resource-manager schedules the radar timeline in time-multiplexing mode.
In the following, we simulate an airspace-surveillance radar rotating at 180°/s with electronic beam-steering.
The airspace is discretised depending on the beam width. Let
where the sensor is located in center of the coordinate system and
Here the factor cos(
To be able to estimate the position of a target continuously in time all radar detections of a target
A more complex dynamic model was introduced by Singer . The state
with discrete white noise
Recursive Bayesian estimators can be used to calculate the state
the mapping between the state space ℒ and the measurement in azimuth
The EKF consists of two steps. First the state
In the general case the matrix
Second the prediction will be
with observation matrix
and Kalman gain
Process noise and measurement error accumulate over time until a new measurement is executed. This leads to a probability density of the track
The probability density is used to calculate the maximum time difference
denotes a projection into the plane orthogonal to the beam direction and
In the simulation presented here a task is generated for each
5.2.4. Performance metrics
In this section three metrics are introduced to validate the performance of the resource manager.
One key element is the tracking accuracy. For the validation the distance between the estimated position
and the real position
To evaluate the surveillance performance, the revisit time is considered. Let therefore be
In this section the validation results of the simulation are presented. The actual airspace situation is depicted in Figure 13. Figures 16 and 17 show the evaluations of the metrics defined in Section 5.2.4.
The simulation starts with an occupied airspace. This can be a difficult situation for the radar since pop-up targets significantly decrease the reaction time as the distance to the radar is shortened.
Figure 16(a) shows that the revisit time for the surveillance settles around a constant value after 4 seconds. The stepped line in Figure 16(b) shows that all targets are tracked in less than 3 seconds. The second line in (b) shows that the tracking accuracy is poor at the beginning of the simulation since the filters need several measurements to initialize correctly. Figure 17 shows that the revisit time oscillates around 4 seconds and that the tracks are stable during routine operation.
In this section, we discuss knowledge-based behavior of a cognitive radar. As discussed in Section 3, the knowledge-based layer works on structured a-priori knowledge about the application domain and its goals and constraints. Automated planning or optimisation tools can be applied to generate mission-level commands that, for example, control the trajectory of the sensor-carrying platform.
Below, we discuss an illustrative trajectory planning problem for a 6-DOF robotic manipulator arm that carries a UWB sensor able to work in synthetic aperture radar (SAR) mode. Results from a real measurement setup using a ST Robotics R17 robot arm are also shown. The sensor has one transmitter and one receiver in a typical common-offset arrangement (Figure 18).
6.1. Robot trajectory planning
The spatial resolution and processing gain that the system can achieve ultimately depend on the trajectory and velocity profile of the sensor head. The constraints can be modeled as an optimisation problem to obtain a feasible, collision-free trajectory of the end-effector of the manipulator arm in Cartesian coordinates that minimizes observation time.
6.1.1. Sensor characteristics and trajectory constraints
The radar sensor under consideration uses a selectable center-frequency from 3 to 8 GHz and 4 GHz of bandwidth, resulting in 3.75 cm of range resolution. The center frequency can be tuned according to a particular target or propagation environment (ground penetration, through-the-wall imaging, IED inspection…). The horn-type antennas can be rotated to exploit polarization diversity. The sensor is able to operate in stripmap or spotlight SAR modes using linear trajectories. Several parallel trajectories can be combined for 3D imaging. The mobility of the arm could be further exploited to generate non-linear trajectories around a target to obtain a more accurate 3D reconstruction.
In order to obtain a similar resolution in cross-range than in range the trajectory planning must (aim to create at least an aperture of 0.5 to 1.3 times the distance to the target in both dimensions (azimuth and elevation) depending on the center frequency used by the system 3 to 8 GHz respectively).
High resolution imaging can only be achieved with an even higher precision positioning. The 3D-trajectory of the sensor needs to be measured and synchronized with the sensor data. For that purpose, accelerometers and gyroscopes from an attached inertial measurement unit (IMU) are used. The IMU drift is additionally stabilized using the hardware readout of optical encoders of the robot arm joints controlled by step-motors.
Two other important parameters to be considered for the trajectory planning are the optimal size of the scanning area and the sampling requirements. Considering the case of planar acquisition geometries working in stripmap mode, to obtain full resolution imaging of the total area of interest, an additional half beam aperture must be extended in both dimensions.
Another important parameter is related with the sampling requirements of a particular acquisition. The measurement positions in the synthetic radar aperture require a minimum spacing in order to sample adequately the phase history associated with all the scatterers. If the distance between measurements is too large the Nyquist criterion is not fulfilled and artifacts may appear in the reconstructed image.
It must be considered also that signal propagation in dielectric materials (ground, wall) will shrink the wavelengths, and sampling requirements become then even more stringent . A previous estimation of the dielectric permittivity of the propagation media may further optimize the acquisition geometry and the imaging process. Figure 19 shows an example of an image obtained with the robot arm using some reference objects inside a plastic suitcase. The trajectory followed by the sensor has been planned considering the constraints previously mentioned to obtain unaliased high-resolution images of the total area of interest.
In this article, a three-layered cognitive radar-architecture based on the Rasmussen model was presented. Several examples illustrated technologies to implement the cognitive subfunctions in a radar system.
For the skill-based layer, an approach for matching a waveform to the target transfer function was shown. In addition, spectrum sensing methods can be used to adapt the transmit signal to the electromagnetic environment. Rule-based behavior can be implemented using Markov-decision processes (MDPs) to compute optimal illumination policies. For a shared-aperture multifunctional radar, radar-resource management approaches are required to schedule the radar timeline. For knowledge-based behavior, an example for sensor-controlled trajectory generation of a robotic-arm were presented.
The different layers of the architecture encompass a broad range of time-scales and levels of abstraction. The full potential is achieved, if all layers interact consistently. This and further experimental validation of the approach are currently investigated at FHR.
Haykin S. Cognitive radar: A way of the future. Signal Processing Magazine. USA: IEEE. Jan. 2006; 23(1):30-40
Guerci JR. Cognitive Radar: The Knowledge-Aided Fully Adaptive Approach. Artech House, Incorporated, 2010. ISBN: 9781596933651
Ender J, Brüggenwirth S. Cognitive radar-enabling techniques for next generation radar systems. In: 2015 16th International Radar Symposium (IRS). IEEE. 2015. pp. 3-12
Rasmussen J. Skills, rules, and knowledge; signals, signs, and symbols, and other distinctions in human performance models. Systems, Man and Cybernetics, IEEE Transactions on SMC. May 1983; 13(3):257-266
Schulte A. Kognitive und kooperative Automation zur Führung unbemannter Luftfahrzeuge. Online Journal Kognitive Systeme. 2013; 1
Sheridan TB. Telerobotics, Automation, and Human Supervisory Control. Cambridge MA: MIT Press; 1992
Wohler M, Schulte A. On board assistant system for UAS integration into civil airspace. In: AIAA Infotech@ Aerospace (I@ A) Conference. 2013. p. 5047
Onken R, Schulte A. System-Ergonomic Design of Cognitive Automation. Heidelberg:Springer; 2010
Putzer H, Onken R. COSA–A generic cognitive system architecture based on a cognitive model of human behavior. Cognition, Technology & Work. 2003; 5(2):140-151
Brüggenwirth S, Schulte A. COSA2 cognitive system architecture with centralized ontology and specific algorithms. In: 2012 International Conference on Systems, Man, and Cybernetics (IEEE SMC 2012). IEEE. 2012
Gjessing DT. Environmental remote sensing, part 1: Methods based on scattering and diffraction of radio waves. Physics in Technology. 1979; 10:266-271
Bell MR. Information theory and radar waveform design. Proceedings. IEEE International Symposium on Information Theory. Jan 1993:428-428. DOI: 10.1109/ISIT.1993. 748744
Schreier PJ, Scharf LL. Statistical Signal Processing of Complex-Valued Data: The Theory of Improper and Noncircular Signals. Cambridge: Cambridge University Press; 2010
Wysocki T, Razavi H, Honary B. Digital Signal Processing for Communication Systems. New York: Springer US; 1997 9780792399322
An Assessment of the Near Term Viability of Accommodating Wireless Broadband Systems in the 1675–1710 MHz, 1755, 1780 MHz, 3500–3650 MHz, and 4200–4220 MHz, 4380–4400 MHz Bands. Tech. rep. U.S. Department of Commerce; 2010
Use of the Bands 5150–5250 MHz, 5250–5350 MHz and 5470–5725 MHz by the Mobile Service for the Implementation of Wireless Access Systems Including Radio Local Area Networks. Tech. rep. ITU-R Resolution 229; 2012
Axell E, et al. Spectrum sensing for cognitive radio. IEEE Signal Processing Magazine. 2012; 29(3):101-116
Yucek T, Arslan H. A survey of spectrum sensing algorithms for cognitive radio applications. IEEE Communications Surveys & Tutorials. 2009; 11(1):116-130
Tandra R, Sahai A. SNR walls for signal detection. IEEE Journal of Selected Topics in Signal Processing. 2008; 2(1):4-17. DOI: 10.1109/JSTSP.2007.914879
Gardner W. Measurement of spectral correlation. IEEE Transactions on Acoustics, Speech, and Signal Processing. 1986; 34(5):1111-1123
Dandawate AV, Giannakis GB. Statistical tests for presence of cyclostationarity. IEEE Transactions on Signal Processing. 1994; 42(9):2355-2369
Zeng Y, Liang Y-c. Eigenvalue-based Spectrum sensing algorithms for cognitive radio. IEEE Transactions on Communications. 2009; 57(6):1784-1793
Stinco P, Greco MS, Gini F. Spectrum sensing and sharing for cognitive radars. IET Radar, Sonar & Navigation. 2016; 10(3):595-602
Kovarskiy JA, et al. Comparison of RF spectrum prediction methods for dynamic spectrum access. In: KI Ranney and A Doerry, ed. Proc. SPIE 10188. May 2017, p. 1018819. DOI: 10.1117/12.2262306
Tian Z, Giannakis GB. A wavelet approach to wideband spectrum sensing for cognitive radios. In: 2006 1st International Conference on Cognitive Radio Oriented Wireless Networks and Communications. June 2006. pp. 1-5. DOI: 10.1109/CROWNCOM.2006.363459
Martone A, et al. Passive sensing for adaptable radar bandwidth. In: 2015 IEEE Radar Conference (RadarCon). May 2015. pp. 0280-0285. DOI: 10.1109/RADAR.2015.7131010
Kirk BH, et al. Cognitive software defined radar: Waveform design for clutter and interference suppression. In: Ranney KI, Doerry A, editors. SPIE Defense + Security. SPIE Proceedings. SPIE. 2017. p. 1018818
Bengio Y. Learning deep architectures for AI. Foundations and Trends in Machine Learning. 2009; 2(1):1-127. ISSN: 1935-8237. DOI: 10.1561/2200000006
Haykin S. Neural Networks and Learning Machines. 3rd ed. Upper Saddle River, New Jersey: Pearson Education Inc.; 2009
Zhao Q, et al. Synthetic aperture radar automatic target recognition with three strategies of learning and representation. Optical Engineering. 2000; 39(5):1230-1244. DOI: 10.1117/1.602495
leCun Y. Generalization and Network Design Strategies. Technical Report CRG-TR-89-4. Department of Computer Science. Toronto, Canada: University of Toronto; June 1989
LeCun Y, et al. Gradient-based learning applied to document recognition. Proceedings of the IEEE. Nov. 1998; 86(11):2278-2324. ISSN: 0018-9219. DOI: 10.1109/5.726791
Schachter BJ. Target classification strategies. Proceedings of SPIE. 2015; 9476:947602-947602. DOI: 10.1117/12.2176193
Lauer F, Bloch G. Incorporating prior knowledge in support vector machines for classification: A review. Neurocomputing. 2008; 71(7–9):1578-1594 ISSN: 0925-2312
Diemunsch JR, Wissinger J. Moving and stationary target acquisition and recognition (MSTAR) model-based automatic target recognition: Search technology for a robust ATR. Proceedings of SPIE. 1998; 3370:481-492. DOI: 10.1117/12.321851
Wagner S, Barth K, Brüggenwirth S. A deep learning SAR ATR system using regularization and prioritized classes. In: 2017 IEEE Radar Conference (RadarConf). May 2017. pp. 0772-0777. DOI: 10.1109/RADAR.2017.7944307
Vapnik VN. The Nature of Statistical Learning Theory. Second. Statistics for Engineering and Information Science. New York: Springer-Verlag; 2000
Wagner SA. SAR ATR by a combination of convolutional neural network and support vector machines. IEEE Transactions on Aerospace and Electronic Systems. Dec. 2016; 52(6):2861-2872. DOI: 10.1109/TAES.2016.160061
Haykin S. Cognitive Dynamic Systems: Perception-Action Cycle, Radar and Radio. Cambridge: Cambridge University Press; 2012. 9780521114363
Castanon DA. Approximate dynamic programming for sensor management. In: Decision and Control, 1997. Proceedings of the 36th IEEE Conference on. Vol. 2. IEEE. 1997. pp. 1202-1207
Blackman S, Popoli R. Design and Analysis of Modern Tracking Systems. Norwood, MA: Artech House; 1999
Singer RA. Estimating optimal tracking filter performance for manned maneuvering targets. IEEE Transactions on Aerospace and Electronic Systems. 1970; 4:473-483
Van Trees HL. Detection, Estimation, and Modulation Theory. New York, NY: John Wiley & Sons; 2004
Grasmueck M, Weger R, Horstmeye H. Full resolution 3-D GPR imaging. Geophysics. 2005; 70(1):12-19
- The training of neural networks is separated into epochs, in each epoch the complete dataset is presented one time to the classifier .