Nanomedicine, Volume I: Basic Capabilities
© 1999 Robert A. Freitas Jr. All Rights Reserved.
Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999
7.1 Nanorobot Communications Requirements
Communication is an important fundamental capability of medical nanorobots. At the most basic level, nanomachines must pass sensory and control data among internal subsystems to ensure stable and correct device operation. They must also exchange messages with biological cells, communicating with the human body at the molecular level. Nanodevices must be able to communicate with each other in order to:
1. coordinate complex, large-scale cooperative activities,
2. pass along relevant sensory, messaging, navigational, and other operational data, and
3. monitor collective task progress.
Finally, nanorobots must be able to receive messages from, and transmit messages to, both the human patient and external entities including antennas and telecommunications links, laboratory or bedside computers, and attending medical personnel.
It is instructive first to examine the gross information flows likely to be required in typical nanomedical situations. Nanorobot onboard computers are likely to generate from 'I ~ 104 bits/sec for the simplest systems to ~109 bits/sec for the most complex systems and tasks (Chapter 12), which sets broad limits on internal communications requirements. Assuming onboard data storage of 105-109 bits, rewriting an entire nanorobot memory in ~1 second demands information flows of 105-109 bits/sec. Communications with human cells may take place at many levels, ranging from 10-1000 bits/sec for individual neuronal impulses (Section 4.8.6) up to brief bursts at ~106 bits/sec (~MHz frequencies) for intracellular processes involving enzyme action or molecular gating. Inter-nanorobot communications are unlikely to require data transfer rates exceeding 103-106 bits/sec; explicit exchanges between nanodevices and the human user are restricted by the maximum data processing rate of the conscious mind, variously estimated as 1-1000 bits/sec (Chapter 25), although dermal and retinal displays may transfer visual information to the patient at up to ~107 bits/sec (Section 7.4.6). Communications with external entities may include monitoring or data transfer operations. For example, the ~525 x 390 = 204,750 pixel/frame of standard halftone black-and-white broadcast television transmitted at 30 frames/sec with ~1 bit/pixel mandates a ~6.1 x 106 bit/sec (6.1 MHz) digital transfer rate (~6.1 MHz transmission bandwidth)*; downloading an entire uncompressed human genome during cytogenetic repair operations (Chapter 20) in ~1000 sec requires a ~107 bits/sec transfer rate.
* Although treated as approximately equal for simplicity in this book, the digital transfer rate is distinct from the transmission bandwidth. The bandwidth needed to handle a specific transfer rate is related to the actual coding and modulation technique selected. The simplest modulation techniques can indeed approach ~1 bit/sec per Hz of bandwidth, but much more efficient techniques are available. For instance, a typical 1998-vintage PC modem used complex coding and modulation schemes to transmit 33 KB/sec over a voice-grade transmission line that is only 4 KHz wide, thus achieving a transfer rate of ~8 bits/sec per Hz of bandwidth; ~100:1 data compression algorithms are currently available for voice transmissions.
According to classical information theory for channel capacity on a dissipative transmission line with additive equilibrium thermal noise,699 the minimum erasure energy required per transmitted bit is
where k = 0.01381 zJ/K (Boltzmann constant), T = 310 K, and 1 zeptojoule (zJ) = 10-21 joule. Thus the maximum 'Imax ~ 109 bits/sec bandwidth requirement noted earlier must draw >~3 picowatts (pW), well within the anticipated 1-1000 pW power budget of typical in vivo medical nanodevices (Section 6.5.3). Slower bit rates can draw even less power. The design challenge is to closely approach this minimum theoretical limit.*
* Landauer700 points out that in theory there is no minimum energy required to transfer a bit; but see also2318 and the discussion of reversible computing in Section 10.2.4.1.
In this Chapter, an analysis of the most common nanorobotic communication modalities (Section 7.2) and communication network architectures (Section 7.3) is followed by a brief discussion of the many specific communication tasks to be performed (Section 7.4).
Last updated on 18 February 2003