Deep Multi-Task Learning for Cooperative NOMA: System Design and Principles

In this work, resource allocation of Cooperative Power Domain Non-Orthogonal Multiple Access (PD-NOMA) with energy harvesting technique using Multiple Input Single Output (MISO) is examined. This paper focuses on improvement of bit error rate (BER), spectral efficiency and capacity compared to conventional Non-Orthogonal Multiple Access (NOMA). Two users scenario is considered, where an user is near to the base station (BS) while the other one is far away from the base station. As both users with different power levels exist in the same spectrum, co-channel interference is inevitable. To alleviate this, an interference cancellation technique known as Successive Interference Cancellation (SIC) is used. The simulation results of cooperative NOMA with energy harvesting are compared with conventional NOMA and it can be inferred that cooperative NOMA performance is better than NOMA.

In non-orthogonal multiple access (NOMA) systems, serving multiple users in shared resource blocks can allow untrusted users to overhear the messages of other users. In this context, we study a network consisting of a base station (BS), a near user and a far user, where the latter attempts to overhear the message of the former. The near user is a full-duplex (FD) node that can also act as a relay. Two operating scenarios are considered: 1) friendly jammer (FJ), where the FD node broadcasts noise for degrading the channel between the BS and the far user, while receiving data from the BS; and 2) friendly jammer relay (FJR), where, in addition to degrading the channel between the BS and the far user, the FD node relays the message of the far user. We investigate the secrecy performance of the network by characterizing the secrecy outage probability (SOP) in both scenarios. We obtain the exact SOP of the FJ case, and an approximation for the SOP of the FJR scenario, both expressed in closed-form. Numerical results confirm the accuracy of the analytical results. For a given BS power budget and power allocation to the users, it is demonstrated that the jamming and relaying powers are prominent factors to make the NOMA-FJR scheme superior to NOMA-FJ, as well as to conventional and cooperative NOMA schemes.

Cooperative non-orthogonal multiple access (NOMA) is heavily studied in the literature as a solution for 5G and beyond 5G applications. Cooperative NOMA transmits a superimposed version of all users’ messages simultaneously with the aid of a relay, after that, each user decodes its own message. Accordingly, NOMA is deemed as a spectral efficient technique. Another emerging technique exploits orbital angular momentum (OAM), where OAM is an attractive character of electromagnetic waves. OAM gathered a great deal of attention in recent years (similar to the case with NOMA) due to its ability to enhance electromagnetic spectrum exploitation, hence increasing the achieved transmission throughput. However, OAM-based transmission suffers from wave divergence, especially at high OAM orders. This OAM limitation reduces the transmission distance. The distance can be extended via cooperative relays (part of cooperative NOMA). Relay helps the source to transmit packets to the destination by providing an additional connection to handle the transmission and provide a shorter distance between source and destination. In this paper, we propose employing OAM transmission in the cooperative NOMA network. Simulation experiments show that OAM transmission helps cooperative NOMA in achieving higher throughput compared to the conventional cooperative NOMA. Concurrently, the cooperation part of cooperative NOMA eases the divergence problem of OAM. In addition, the proposed system outperforms the standalone cooperative OAM-based solution.

The full-duplex (FD) cooperative nonorthogonal multiple access (NOMA) achieves a superior throughput over the conventional half-duplex (HD) cooperative NOMA, wherein the strong users (SUs) with good channel conditions can act as an FD relay node for the weak users with poor channel conditions. However, the energy efficiency (EE) of a cooperative NOMA may be degraded due to the additional power consumption incurred at the SUs. We are, therefore, motivated to investigate the EE maximization problem of an FD cooperative NOMA system. More importantly, we investigate the “signal-to-inference-noise ratio gap reversal” problem of cooperative NOMA systems, which imposes successive interference cancellation (SIC) performance degradation at the SUs. This problem has not been documented in the existing cooperative NOMA literature. A low-complexity algorithm is proposed for maximizing the system's EE while guaranteeing a successful SIC operation. Our numerical results show that the proposed algorithm achieves both a higher EE and throughput over the existing HD cooperative NOMA and nonadaptive FD cooperative NOMA. More importantly, the proposed scheme guarantees a successful SIC operation at the SUs.

In this article, we investigate max completion time optimization for Internet of Things (IoT) in LEO satellite-terrestrial integrated networks (STINs), in which IoT devices use non-orthogonal multiple access (NOMA) scheme to transmit data to central earth stations (CESs), and orthogonal multiple access (OMA) scheme is used for data transmission from CESs to LEO satellite. We decouple this problem into two subproblems: 1) max completion time optimization in terrestrial networks and 2) max completion time optimization among satellite beams. Different from the existing works about NOMA data transmission in terrestrial networks, we propose a cooperative NOMA scheme, and derive the closed expressions of the optimal cooperative data and the optimal transmit power of IoT devices. Based on the closed-form expressions, a joint subcarrier assignment and cooperative NOMA pairing (JSACNP) approach is proposed to minimize the max completion time in terrestrial networks by utilizing matching theory. Then, to minimize the max completion time among satellite beams, the optimal linear receiver expression is derived with fixed transmit power. Convex optimization is utilized to solve transmit power optimization, we propose an algorithm to solve it by CVX tool. An iterative algorithm is proposed for improved performance. Finally, numerical results are provided to evaluate our proposed algorithms, compared with some other proposed approaches or algorithms.

Cooperative communication is one of the emerging technologies that promise significantly higher reliability and spectral efficiency in wireless networks. It is a new form of diversity that allows users or nodes to share resources to create collaboration via distributed transmission and processing of messages. As candidate multiple access for future radio access, Non orthogonal multiple access (NOMA) over orthogonal multiple access (OMA) are adopted as downlink access scheme for future 5th generation mobile communication. In this paper, we compare user throughput when using orthogonal multiple access (OMA) and NOMA in macro femto network. We show through computer simulation the effect of imperfect interference cancellation on user throughput. We assist the performance of user selection when using fixed power allocation. Finally, we propose NOMA with COMP scenario whose performance improvement lead for user and cell throughput.

The performance of full-duplex (FD) cooperative non-orthogonal multiple access (C-NOMA) relaying systems in the presence of in-phase and quadrature-phase imbalance (IQI) and imperfect successive interference cancellation (ipSIC) is analyzed and evaluated. In particular, by considering two classes of users located far from and near to the information source, closed-form exact and analytical approximate expressions for the outage probability (OP) and ergodic sum rate (ESR), respectively, are derived. The accuracy of the analytical approach is verified by computer simulations which have shown that for both C-NOMA and equivalent OMA systems IQI has a deleterious effect on their OP and ESR performance in moderate and high signal-to-noise ratio (SNR) regions and the OP of C-NOMA is always slightly lower than that of the traditional OMA. Additionally, comparisons with equivalent half-duplex (HD) C-NOMA schemes have revealed that for the far user the FD C-NOMA scheme is less sensitive to ipSIC.

Geolocation of mobile nodes with non‐orthogonal multiple access (NOMA) is a crucial process for enabling advanced location‐based services (LBSs) in 5G networks. In this study, the authors investigate the theoretical limits of received‐signal‐strength (RSS)‐based positioning for uplink cooperative NOMA under different network conditions with special emphasis on the user fairness feature. In addition, they compare the performance of the RSS‐based positioning of cooperative NOMA with that of orthogonal multiple access (OMA). The Cramer‐Rao lower bounds are derived in order to show the fundamental limits of position estimation error by taking amplify‐and‐forward and decode‐and‐forward relaying into account. The numerical results for both schemes are presented and compared considering the effects of number of relays and user equipments, different relaying on the performance of RSS positioning. The effects of power allocation on the user fairness of cooperative NOMA are also investigated. The results show that the cooperative NOMA outperforms cooperative OMA in terms of positioning. It provides high precision position accuracy and user fairness under different network conditions. Hence, the results suggest that the RSS‐based positioning for uplink cooperative NOMA is a promising technique for achieving user fairness and supporting diversified low‐cost LBSs in 5G networks.

Grant-free Non-Orthogonal Multiple Access (NOMA) is a promising solution to enable massive wireless access service for 5G systems and beyond. Conventional grant-free NOMA schemes directly apply the spreading signatures optimized for the grant-based scenarios, which, however, ignore the users' diversified activation probabilities. In addition, the conventional grant-free NOMA schemes are not designed with finite-alphabet signatures, which hinders the encoder/decoder implementation using practical low-cost hardware. Therefore, to overcome these limitations, we propose a finite-alphabet signature design for the grant-free NOMA with random and nonuniform user activations. Herein, the NOMA signatures are optimized by the autoencoder-based transceivers with both transmitter and receiver being in the form of deep neural network. First, the quantized deep learning is employed in the NOMA signature training, which jointly optimizes the sequence generation and quantization. Moreover, in order to improve the training rate, we propose a specific neural network receiver, where the network structure resembles the successive interference cancellation procedures. The experiment results show that the obtained NOMA signatures commendably exploit the users' activation profiles, and the proposed scheme outperforms the conventional ones especially when the users have unequal activation probabilities.

This paper has investigated a downlink non-orthogonal multiple access (NOMA) cooperative scheme with user-assisted relaying, where near users are viewed as decode-and-forward (DF) relays operating full/half-duplex mode to assist multiple far users. Especially, the impact for the randomness of users location on the system performance has also been studied. Further, to overcome the zero diversity gain inherent to full-duplex (FD) operation, the direct link between a base station (BS) and each far user is exploited. The comprehensive performance analysis is conducted through the closed-form expressions in terms of outage probability, the diversity gain, system throughput, and ergodic sum rate. With respect to outage behavior, throughput, and ergodic sum rate, FD-based NOMA outperforms half-duplex (HD) NOMA in the lower signal-to-noise ratio (SNR) region. However, with the negative influence brought by residual loop-interference (LI), the advantage of FD NOMA scheme is not pronounced enough in the high SNR regime. Therefore, according to different SNR levels, we further propose a hybrid relaying scheme that switches opportunistically between the FD and HD modes. Afterward, the mode switching threshold is characterized such that the outage probability of each user is minimized. The simulation results confirm the correctness of our analysis and offer a significant guideline for system design in practice.

Recent developments in technologies like the internet of Things (ioT) and massive machine-type communications (mMTC), the number of wireless communication devices, the demand for spectrum resources, the power consumed in transmission, processing, and reception will proliferate. Non-orthogonal multiple access (NOMA) and cooperative schemes are attracting researchers across the globe to use NOMA to solve the problem of spectrum scarcity by serving multiple users simultaneously, whereas cooperative schemes can help in reducing transmission power consumption by employing relay nodes in between. Therefore, to condense the system's power consumption, this article integrates NOMA and cooperative technology. This article proposes two schemes: a battery efficient cooperative NOMA scheme and a customized sector model. The introduced battery efficient cooperative NOMA takes advantage of relay and device-to-device (D2D) cooperation for introducing different modes of operation in the NOMA scheme, aiming to reduce the battery power consumption at the transmitter and the receiver side, whereas the customized sector model intends to optimize the energy of the system through efficient deployment of the NOMA cell-sector model. Finally, the discussion of various challenges and future opportunities for unveiling the NOMA technique is presented.

The non-orthogonal multiple access (NOMA) is one of the fledging paradigms which next generation radio access technologies are sprouting toward. The NOMA with superposition coding (SC) in the transmitter and successive interference cancellation (SIC) at the receiver comes with many desirable features and benefits over orthogonal multiple access (OMA) such as orthogonal frequency division multiple access (OFDMA) adopted by Long-Term Evolution (LTE). In this paper, we study the recent research trends on NOMA in 5G systems. We discuss the basic concept of NOMA and explain its aspects of importance for future radio access. Then, we provide a survey of the state of the art in NOMA for 5G systems in a categorized manner. Further, we analyze the NOMA performances with numerical examples; and provide some avenues for future research on NOMA on a set of open issues and challenges.

Due to the successive interference cancellation (SIC) technique, Non orthogonal multiple Access (NOMA) users have the capability of decoding the other paired users sharing the same resource bloc. As a result, NOMA technique is not only exposed to external eavesdropping, but also to internal eavesdropping which raises security risks regarding NOMA users' privacy. Therefore, this research aims to analyse a Multi-antenna cooperative NOMA system where both Decode and Forward (DF) and Amplify and Forward (AF) strategies are studied. In this context, we investigate the secrecy performance of our system with both internal and external eavesdropping scenarios. Closed-form expressions of the secrecy outage probability (SOP) and the strictly positive secrecy capacity (SPSC) metrics are derived. Numerical results are given to illustrate the validity of the obtained theoretical results.

This paper investigates the application of simultaneous wireless information and power transfer (SWIPT) to cooperative non-orthogonal multiple access (NOMA). A new cooperative multiple-input single-output (MISO) SWIPT NOMA protocol is proposed, where a user with a strong channel condition acts as an energy-harvesting (EH) relay to help a user with a poor channel condition. The power splitting (PS) scheme is adopted at the EH relay. By jointly optimizing the PS ratio and the beamforming vectors, the design objective is to maximize the data rate of the "strong user" while satisfying the QoS requirement of the "weak user". It boils down to a challenging nonconvex problem. To resolve this issue, the semidefinite relaxation (SDR) technique is applied to relax the quadratic terms related with the beamformers, and then it is solved to its global optimality by two-dimensional exhaustive search. We prove the rank-one optimality, which establishes the equivalence between the relaxed problem and the original one. To further reduce the high complexity due to the exhaustive search, an iterative algorithm based on successive convex approximation (SCA) is proposed, which can at least attain its stationary point efficiently. In view of the potential application scenarios, e.g., IoT, the single-input single-output (SISO) case of the cooperative SWIPT NOMA system is also studied. The formulated problem is proved to be strictly unimodal with respect to the PS ratio. Hence, a golden section search (GSS) based algorithm with closed-form solution at each step is proposed to find the unique global optimal solution. It is worth pointing out that the SCA method can also converge to the optimal solution in SISO cases. In the numerical simulation, the proposed algorithm is numerically shown to converge within a few iterations, and the SWIPT-aided NOMA protocol outperforms the existing transmission protocols.

The advancement in deep learning has taken the technology world by storm in the last decade. Although, there is enormous progress made in terms of algorithm performance, the security aspect of these algorithms has not received a lot of attention from the research community. As more industries start to adopt these algorithms the issue of security is becoming even more relevant. Security vulnerabilities in machine learning (ML), especially in deep neural networks (DNN), is becoming a concern. Various techniques have been proposed, including data manipulations and model stealing. However, most of them are focused on ML algorithms and target threat models that require access to training dataset. In this paper, we present a methodology that analyzes the DNN weight parameters under the threat model that assumes the attacker has the access to the weight memory only. This analysis is then used to develop an attack that manipulates weight parameters with respect to their sensitivity. To evaluate this attack, we implemented our methodology on a MLP trained on IRIS dataset and LeNet (DNN architecture) trained on MNIST dataset. Our experimental results demonstrate that alteration of model parameters results in subtle accuracy drop of the model. Depending on the applications such subtle changes can cause significant system malfunction or disruption, for example in vision-based industrial applications. Our results show that using our methodology a subtle accuracy drop can be achieved in a reasonable amount of time with very few parameter changes.