Ehab Mahmoud Mohamed

On-Demand Hybrid Routing for Cognitive Radio Ad-Hoc Network

A novel on-demand cluster-based hybrid routing protocol for cognitive radio ad hoc network with non-uniform node distribution is proposed in this paper. At first, a novel spectrum-aware clustering mechanism is introduced. The proposed clustering mechanism divides node into clusters based on three values: spectrum availability, power level of node, and node stability. Therefore, clusters are formed with the highest stability to avoid frequent reclustering. Later, a routing algorithm is introduced to minimize the delay while achieving acceptable delivery ratio. In this paper, routing is defined as a multi-objective optimization problem to combine different individual routing metrics to form a global metric. Simulation results show that our proposed routing algorithm can guarantee a lower delay and a higher packet delivery ratio than conventional routing protocols for cognitive radio ad hoc networks. Due to our routing design specification, such as power consideration and low delay, it can be suitable to be adopted for Internet of Things applications.

Delayed offloading zone associations using cloud cooperated heterogeneous networks

Delayed offloading of mobile user delay tolerant traffic, through other networks such as WiFi networks or millimeter wave (mm-w) gates, is a promising solution for dealing with the ever-growing problem of wireless cellular network capacity. In this paper, we investigate the problem of how we can optimally associate a user with delayed traffic to a nearby offloading zone, i.e., to which nearby offloading zone a user with delayed traffic should be associated. Towards that, we propose an adaptive scheme for offloading zone association, including a weighted proportional fairness (WPF) online algorithm for optimal zone selection. The proposed scheme is a network-based, which maximizes the total system offloading and the users offloading experiences (OFEs). To efficiently implement the proposed scheme, we also suggest a cloud cooperated heterogeneous network (CC-HetNet) to hold the delayed offloading process in a cloud cooperated manner. In this CC-HetNet, the offloading zones and the Macro Base station (BS) are linked to the Centralized Radio Access Network (C-RAN) using high speed backhaul links. The proposed scheme of offloading zone association will be implemented in the C-RAN as an enhanced Access Network Discovery and Selection Function for delayed offloading networks.

A complexity efficient equalization technique for MIMO-constant envelope modulation

MIMO-Constant Envelope Modulation, MIMO-CEM, with Low Resolution ADC receiver (1-bit ADC in the default operation) is considered as one of the alternative candidates to the well-known MIMO-OFDM, suitable for wireless backhaul networks. Because MIMO-CEM receiver is based upon low resolution ADC sampled at the Intermediate Frequency (IF) band, the currently used MIMO-CEM equalizer is too complicated. The equalizer high complexity comes from its design; in which IF based Maximum Likelihood Sequence Estimator (MLSE) is utilized. In order to efficiently estimate the transmitted sequence, the MLSE equalizer replicates the whole IF MIMO-CEM transceiver, which results in too complicated MIMO-CEM equalizer. In this paper, low complexity equalization technique for MIMO-CEM is proposed. In the proposed method, a baseband approximation model for the 1-bit ADC MIMO-CEM is investigated. Through using this model, a baseband based MIMO-CEM MLSE equalizer can be constructed. This baseband equalizer will nominate number of candidate sequences, which are further refined by the IF based MLSE equalizer to accurately estimate the transmitted sequence. Hence, the number of equalization processes required by the high complexity IF based MLSE are greatly reduced, which powerfully contributes in reducing the complexity of MIMO-CEM equalization. The effectiveness of the proposed estimator is proven under different scenarios using Minimum Shift Keying (MSK) and Gaussian MSK (GMSK) modulations.

Affine linear transformation based sphere decoder for 1-bit ADC MIMO-constant envelope modulation

Multi-input multi-output constant envelope modulation (MIMO-CEM) is considered as a promising low power PHY layer design for emerging Internet of Things (IoT) smart devices. MIMO-CEM efficiently overwhelms the high power consumption in MIMO-OFDM transmit signal comes from its high peak-to-average power ratio (PAPR). Despite its high efficiency in reducing the power consumption in the radio frequency (RF) band, MIMO-CEM suffers from high computational complexity in the digital signal processing performed by the receiver, e.g., MIMO decoding and channel estimation. This comes from the 1-bit analog to digital converter (ADC) operating in the intermediate frequency (IF) band used at the receiver side. In this paper, a low complexity sphere decoder (SD) is proposed for MIMO-CEM. A low complexity baseband modeling of MIMO-CEM based on an affine linear transformation is used to specify the sphere radius in addition to enumerating the lattice candidates within the sphere. Then, the more accurate (the more complex) IF based maximum likelihood decoder (MLD) is used to pick up the candidate that most likely to be transmitted. Decreasing the number of candidates examined by the high complex IF based MLD leads to a great complexity reduction in the proposed decoder. We test the proposed MIMO-CEM decoder under different scenarios and compare its performance with the recent decoders proposed for MIMO-CEM detection.

Millimeter wave beamforming training, discovery and association using WiFi positioning in outdoor urban environment

Millimeter wave (mmWave) communication is considered as the most promising solution for future 5G cellular networks. However, it suffers from a severe propagation loss causing dramatic link quality degradation and limiting its coverage area. Antenna beamforming appears to be a suitable solution to overcome this problem and provides longer transmission outdoor mmWave links. Accordingly, highly complicated conventional beamforming training (BT) techniques based on exhaustively searching all beam settings in all possible directions are used for mmWave discovery and association. This results in a long-time delay and a high-power consumption mmWave small cell discovery and association due to the large BT overhead. In this paper, a paradigm of mmWave small cell BT, discovery and association based on WiFi localization is proposed especially for local outdoor urban environments, e.g., campus area. Due to the use of beamforming, mmWave becomes a location-driven wireless communication system, hence positioning can extremely relax the complexity of mmWave link establishment. Simulation results show the high potency of the proposed scheme in reducing the complexity of mmWave BT, discovery and association compared to the conventional scheme of exhaustively searching all beam directions.

Millimeter wave location-based beamforming using compressive sensing

This paper develops a location based analog beamforming (BF) technique using compressive sensing (CS) to be feasible for millimeter wave (mmWave) wireless communication systems. The proposed scheme is based on exploiting the benefits of CS and localization to reduce mmWave beamforming (BF) complexity and enhance its performance compared with conventional mmWave analog BF techniques. CS theory is used to exploit the sparse nature of the mmWave propagation channel to estimate both the angle of departures (AoDs) and the angle of arrivals (AoAs) of the mmWave channel, and knowing the node location effectively reduces the number of BF vectors required for constructing the sensing matrix. Hence, a high accurate mmWave BF with a low set-up time can be obtained. Simulation analysis confirms the high effectiveness of the proposed mmWave BF technique compared to the conventional exhaustive search BF and the CS based BF without localization using random measurements.

Low complexity MIMO detection technique for 1-bit ADC MIMO-CEM using adaptive sphere decoding

Multiple input, multiples output (MIMO)-constant envelop modulation (CEM), has been proposed to be an efficient power alternative transceiver system to MIMO-OFDM. To attain a high power efficient MIMO transceiver, MIMO-CEM uses a constant envelope modulation at the transmitter (TX) side and an intermediate frequency (IF) based 1-bit ADC in the receiver (RX) side. Despite MIMO-CEM introduces low power consumption, it suffers from highly complicated MIMO detection due the use of IF-based 1-bit ADC in the RX. A decoder with low complexity is proposed in this paper for MIMO-CEM system in the presence of 1-bit ADC. In which, as a first step, an adaptive sphere decoder (ASD) rely on the affine linear transformation (ALT) model for MIMO-CEM system is proposed to adaptively select a small number of sequences from the all lattice states. Thereafter, the exceedingly accurate (the extra complex) maximum likelihood decoder (MLD) in the IF-band for MIMO-CEM is utilized to estimate the transmit sequence which is the most probable to be sent from the selected sequences. In the suggested adaptive SD scheme, the sphere radius is adaptively adjusted depend on the channel conditions, e.g., signal-to-noise power ratio (SNR). Simulation results confirm the high efficiency of the proposed adaptive SD in reducing the complexity of MIMO-CEM detection compared to other existing techniques.

Experimental work on WiGig coverage area management and beamforming training using Wi-Fi fingerprint

Although Wireless Gigabit (WiGig) access operating in the 60 GHz band plays a significant role towards multi-Gbps WLANs, its transmission suffers from harsh propagation loss and path blocking reducing its transmission range to be few meters around a WiGig access point/station (AP/STA). Consequently, directional transmissions using antenna beamforming is tremendously used in WiGig communication. In this paper, a novel approach of leveraging Wi-Fi channel fingerprints for localizing WiGig coverage area along with reducing its beamforming training (BT) complexity over conventional exhaustive search BT is proposed. The proposed approach is motivated by the fact that Wi-Fi fingerprints, WiGig coverage area and the best beam identifications (IDs) of a WiGig AP and STA are all location dependent. Hence, by linking Wi-Fi fingerprints with WiGig information, e.g., WiGig coverage and the best AP/STA beam IDs, using statistical learning, Wi-Fi fingerprints comparisons can be used to detect if a WiGig STA is within the coverage area of a WiGig AP or not and which AP/STA beam IDs are expected to maximize the link quality. Experimental work in real indoor environment is conducted to prove the effectiveness of the proposed approach compared to the conventional exhaustive search BT.

Low Density Parity Check (LDPC) coded MIMO-Constant Envelop Modulation System with IF sampled 1-bit ADC

MIMO-Constant Envelop Modulation (CEM) is a very power and complexity efficient system, which is introduced as alternative candidate to the currently used MIMO-Orthogonal Frequency Division Multiplexing (OFDM). CEM system enables to use high efficient nonlinear power amplifier on the transmitter side and 1 bit (low resolution) analog to digital converter (ADC) on the receiver side. Due to adopting the low resolution at the receiver side a great reduction in hardware complexity and power consumption can be achieved. However, there will be a noticeable degradation on the performance of bit error rate (BER) on the receiver side due to sever quantization error introduced by the low resolution ADC, so a forward error correction coding is essential to enhance the BER. In this paper a LDPC coded MIMO-CEM system was used as a replacement for MIMO-OFDM to deal with the BER degradation problem of the CEM system. The performance of the LDPC coded MIMO-CEM with Gaussian Minimum Phase Shift Keying (GMSK) modulation is evaluated over a multi-path Rayleigh fading channel. It showed that LDPC codes are effective to improve the BER performance of CEM on Rayleigh fading channels. According to the simulation results, the MIMO-CEM system provides a significant improvement in BER performance and outperforms the un-coded and the original convolutional coder based CEM systems.

Low complexity channel estimation technique for MIMO-Constant Envelope Modulation

Low-resolution ADC MIMO-Constant Envelope Modulation, MIMO-CEM, is one of the alternative candidates to the well-known MIMO-OFDM, suitable for wireless backhaul networks. One of the major problems withstands the real application of MIMO-CEM is the channel estimation. This comes from the low resolution ADC used in MIMO-CEM receiver. Intermediate Frequency (IF) based Adaptive channel estimator was previously proposed for MIMO-CEM. This channel estimator is based upon replicating the received preamble signal to exactly match the effect of high quantization error ADC upon the received MIMO signal. Because MIMO-CEM ADC is IF sampled, the whole IF MIMO-CEM transceiver is replicated to exactly replicate the received preamble signal. Although this channel estimator has an excellent estimation performance, its computational complexity is very high. This high complexity comes from the IF (high frequency) calculations. In this paper, a low complexity MIMO-CEM adaptive channel estimator is introduced. The proposed channel estimator is based upon linearly modeling the quantization error occurred at the low resolution IF sampled ADC in baseband. Hence, a baseband adaptive channel estimator can be designed, which has a much lower complexity than the IF estimator (previously designed). The effectiveness of the proposed estimator is proven under different channel scenarios using Minimum Shift Keying (MSK) and Gaussian MSK (GMSK) modulations.