2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)最新文献

{"title":"Trojan Attacks on Wireless Signal Classification with Adversarial Machine Learning","authors":"Kemal Davaslioglu, Y. Sagduyu","doi":"10.1109/DySPAN.2019.8935782","DOIUrl":"https://doi.org/10.1109/DySPAN.2019.8935782","url":null,"abstract":"We present a Trojan (backdoor or trapdoor) attack that targets deep learning applications in wireless communications. A deep learning classifier is considered to classify wireless signals using raw (I/Q) samples as features and modulation types as labels. An adversary slightly manipulates training data by inserting Trojans (i.e., triggers) to only few training data samples by modifying their phases and changing the labels of these samples to a target label. This poisoned training data is used to train the deep learning classifier. In test (inference) time, an adversary transmits signals with the same phase shift that was added as a trigger during training. While the receiver can accurately classify clean (unpoisoned) signals without triggers, it cannot reliably classify signals poisoned with triggers. This stealth attack remains hidden until activated by poisoned inputs (Trojans) to bypass a signal classifier (e.g., for authentication). We show that this attack is successful over different channel conditions and cannot be mitigated by simply preprocessing the training and test data with random phase variations. To detect this attack, activation based outlier detection is considered with statistical as well as clustering techniques. We show that the latter one can detect Trojan attacks even if few samples are poisoned.","PeriodicalId":278172,"journal":{"name":"2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124727053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning for RF Signal Classification in Unknown and Dynamic Spectrum Environments","authors":"Yi Shi, Kemal Davaslioglu, Y. Sagduyu, W. Headley, Michael Fowler, Gilbert Green","doi":"10.1109/DySPAN.2019.8935684","DOIUrl":"https://doi.org/10.1109/DySPAN.2019.8935684","url":null,"abstract":"Dynamic spectrum access (DSA) benefits from detection and classification of interference sources including in-network users, out-network users, and jammers that may all coexist in a wireless network. We present a deep learning based signal (modulation) classification solution in a realistic wireless network setting, where 1) signal types may change over time; 2) some signal types may be unknown for which there is no training data; 3) signals may be spoofed such as the smart jammers replaying other signal types; and 4) different signal types may be superimposed due to the interference from concurrent transmissions. For case 1, we apply continual learning and train a Convolutional Neural Network (CNN) using an Elastic Weight Consolidation (EWC) based loss. For case 2, we detect unknown signals via outlier detection applied to the outputs of convolutional layers using Minimum Covariance Determinant (MCD) and k-means clustering methods. For case 3, we extend the CNN structure to capture phase shifts due to radio hardware effects to identify the spoofing signal sources. For case 4, we apply blind source separation using Independent Component Analysis (ICA) to separate interfering signals. We utilize the signal classification results in a distributed scheduling protocol, where in-network (secondary) users employ signal classification scores to make channel access decisions and share the spectrum with each other while avoiding interference with out-network (primary) users and jammers. Compared with benchmark TDMA-based schemes, we show that distributed scheduling constructed upon signal classification results provides major improvements to in-network user throughput and out-network user success ratio.","PeriodicalId":278172,"journal":{"name":"2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125684625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Should We Worry About Interference in Emerging Dense NGSO Satellite Constellations?","authors":"C. Braun, Andra M. Voicu, L. Simić, P. Mähönen","doi":"10.1109/DySPAN.2019.8935875","DOIUrl":"https://doi.org/10.1109/DySPAN.2019.8935875","url":null,"abstract":"Many satellite operators are currently planning to deploy non-geostationary-satellite orbit (NGSO) systems for broadband communication services in the Ku-, Ka-, and V-band, where some of them have already started launching. Consequently, new challenges are expected for inter-system satellite coexistence due to the increase in the interference level and the complexity of the interactions resulting from the heterogeneity of the constellations. This is especially relevant for the Ku-band, where the NGSO systems are most diverse and existing geostationary-satellite orbit (GSO) systems, which often support critical services, must be protected from interference. It is thus imperative to evaluate the impact of mutual inter-system interference, the efficiency of the basic interference mitigation techniques, and whether regulatory intervention is needed for these new systems. We conduct an extensive study of inter-satellite coexistence in the Ku-band, where we consider all recently proposed NGSO and some selected GSO systems. Our throughput degradation results suggest that existing spectrum regulation may be insufficient to ensure GSO protection from NGSO interference, especially due to the high transmit power of the low Earth orbit (LEO) Kepler satellites. This also results in strong interference towards other NGSO systems, where traditional interference mitigation techniques like took-aside may perform poorly. Specifically, took-aside can be beneficial for large constellations, but detrimental for small constellations. Furthermore, we confirm that band-splitting among satellite operators significantly degrades throughput, also for the Ku-band. Our results overall show that the complexity of the inter-satellite interactions for new NGSO systems is too high to be managed via simple interference mitigation techniques. This means that more sophisticated engineering solutions, and potentially even more strict regulatory requirements, will be needed to ensure coexistence in emerging, dense NGSO deployments.","PeriodicalId":278172,"journal":{"name":"2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124363914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}