Use of magnetic fields for underwater wireless communication recently has been found very promising due to its advantages of being low power, low cost, no multipath, no Doppler effect and instant communication. Keeping in view the importance and the potential of magnetic field communication for underwater wireless sensor networks, this paper presents a review of the fundamental concepts and theory of magnetic field communication by identifying the main blocks, describing the limitation of each block and mentioning the advancements being done. The three fundamental blocks identified are: 1) the transmit and receive coils, 2) matching network, and 3) channel characteristics of magnetic fields. The first block explains how a single dimensional coil is used to establish a directional communication link between a transmitter & receiver and then how the directionality problem is solved with the help of multi-dimensional coils and meta-material coils. Further, the matching network, explains the difference between the induction and resonance phenomena and their effect on the range and capacity. In the third block, the factors affecting the magnetic fields are mentioned that are critical for underwater wireless communication with magnetic fields. Finally, the paper introduces the recent advancements at physical, MAC and network layers.

A set of low-cost and energy-efficient sensor nodes are designed and implemented using magnetic induction (MI) communications for wireless sensor networks (WSN), which are especially suited for underwater and underground networks where the conventional mode of communication does not perform well. Hardware features of the sensor nodes include a three-dimensional MI coil antenna and its different configurations for transmit and receive operations, low-power circuits for sleep mode, several types of sensors, data storage, and best transmit/receive circuits selected to achieve the maximum communication range with low-power consumption. The material cost of a sensor node is less than $\$$ 100 USD at the prototyping stage and can be drastically reduced at volume production. Software design utilizes low-power modes of microcontrollers and power supply circuits, state machine implementation of sleep, receive, sensing, and transmit modes, range estimation from received signal strength indicators (RSSIs), and medium access protocols, such as carrier sense multiple access (CSMA). Extensive lab and field tests conducted with the sensor nodes demonstrate promising performance in terms of low-power consumption, communication range, range estimation, and robustness against mismatching of coil orientations, and networking capabilities. The current consumption is as low as 60 $\mu$ A in sleep mode, 0.49 $\text{m}$ A in receive mode, and 253 $\text{m}$ A in transmit mode. The sensor node achieves a maximum of 40-m range with 1 kb/s data rate at 125 kHz carrier frequency.

High-speed underwater acoustic (UWA) communication is challenging due to limited bandwidth, extended multipath spread, severe Doppler effect, and large transmission loss. The use of multiple transducers at both transmitter and receiver increases the spatial diversity of the system, and orthogonal frequency-division multiplexing (OFDM) solves the problem of limited bandwidth due to its high spectral efficiency. Thus, the combination of multiple-input multiple-output (MIMO) and OFDM shifts the UWA communication to a whole new level; however, at the same time, it also increases the design challenges for the development of such systems. Intersymbol interference, intercarrier interference, and cochannel interference are introduced in the received signal by the extended multipath and Doppler shifts along with different types of noises due to the noisy acoustic channel. Therefore, many channel estimation and equalization schemes are developed for the UWA MIMO-OFDM communication systems. In this paper, these schemes are thoroughly reviewed, summarized, categorized, and compared with other similar algorithms in terms of complexity and performance while pointing out the prevailing research problems.

Magneto-inductive (MI) waveguide technology is often proposed to increase the MI communication distance without adding significant cost and power consumption to the wireless sensor network. The idea is to add intermediate relaying nodes between transmitter (Tx) and receiver (Rx) to relay the information from Tx to Rx. Our study of MI wave-guides has realized that adding a relay node improves the communication distance, however, the performance is greatly dependent on the position of the relaying node in the network. We therefore, in this work have investigated the effect of placement of a relay node and have determined the optimal relay position. We have performed various sets of experiments to thoroughly understand the behavior and identified three main regions: (a) for region 1, when the distance between Tx and Rx is equal or less than the diameter of the coils ( $d\le 2r$ ), the optimal relay position is close to Tx, (b) for region 2, when the distance between Tx and Rx is greater than diameter of the coils but less than twice the diameter ( $2r<d<4r$ ), the optimal relay position lies in the center of Tx and Rx, and (c) for region 3, when the distance between the Tx and Rx is equal or greater than twice the diameter of the coils ( $d\ge 4r$ ), the optimal relay position is close to Rx.