Blockchain Papers

Chong Yang, Keli Chen, Zhuhua Hu, Yaochi Zhao · 6 authors

With the rapid development of communication technology, the efficient utilization and management of spectrum resources have become a critical issue for the performance and reliability of wireless communication networks. Collaborative spectrum sensing, as an important technology for enhancing spectrum utilization, can significantly improve the accuracy of spectrum sensing through multi-node cooperation. However, existing methods face challenges such as malicious node behavior, data tampering, and unfair resource allocation in practical applications, thereby severely constraining the performance and security of spectrum sensing systems. In response to these issues, this paper proposes a blockchain-based collaborative spectrum sensing system and introduces two key innovations focusing on spectrum sensing and resource optimization: Firstly, an innovative hybrid consensus mechanism called Proof of Authority and Stake (PoAS) has been designed. It organically combines the efficiency of authoritative nodes with the fairness of stake distribution, thus optimizing the consensus process of the blockchain and ensuring the timeliness and trustworthiness of spectrum sensing data. Secondly, a detection method for malicious users is proposed based on game theory, which identifies and suppresses the behavior of malicious nodes in real-time through a dynamic reputation mechanism, thereby enhancing both the accuracy of spectrum sensing and the robustness of the system. Additionally, this paper constructs a collaborative sensing framework that integrates optimized spectrum resource allocation, striking a balance between sensing performance and resource allocation efficiency. Experimental results indicate that the system proposed in this paper, when compared to the single PoAS hybrid mechanism, can elevate the detection probability from 0.5 to 0.9 through the implementation of the PoAS+ game theory mechanism. Furthermore, the average difference in node revenue is enhanced by 454 units, thereby augmenting the accuracy of spectrum sensing and improving resource utilization.

Jerome I. Rotgans, David M. Boom

The objective of this proof-of-concept study was to test the utility of NeuroTargeted Training (NTT), a new method using functional Near-Infrared Spectroscopy (fNIRS) to measure and enhance cognitive performance during simulator training. Traditional simulator training is limited to behavioral evaluations, without capturing the trainee's internal cognitive processes. NTT addresses this gap by comparing trainees’ brain activation patterns to those of experts, allowing for precise identification and remediation of cognitive performance gaps. Three studies were conducted with five participants. Expert neural benchmarks were established from a man overboard simulation. Novices were evaluated against these benchmarks using a Expert Reference Index (ERI), quantifying deviations from expert performance, and the NeuroTargeted Training methodology was compared with conventional evaluations. Personalized training, based on identified gaps, was conducted to align novice neural patterns with expert benchmarks. Significant differences were observed, particularly in the anterior insula and inferior frontal gyrus, with an ERI of 4.84. Cohen’s Kappa (.69) indicated moderate inter-rater reliability. Subsequent targeted training reduced the ERI by 27%, aligning novice neural patterns with experts. Without intervention, the ERI rose by 79%, indicating increased cognitive strain. These findings highlight NTT’s potential to enhance learning outcomes in high-stake exercises by providing insights into cognitive processes.

Billy R. Hammond, Billy R. Wooten, Bill Smollon

We are pleased to respond to Gellerman and Bernstein's (G&B) critique of our review. Unlike many fields, science is self-corrective and largely succeeds because of its openness to criticism. We applaud G&B's efforts to develop a useful device for the important purpose of measuring macular pigment (MP) noninvasively. We were disappointed, however, that G&B feel that we have an “unyielding bias against the use of Raman spectroscopy to quantify MP in vivo” We would criticize any method that made claims regarding these important issues that were not supported by reasonable evidence. In our review, we critically analyzed both psychophysical and physical methods and noted shortcomings and strengths. We pointed out that “as currently conceived,” the Raman method of MP optical density measurement is invalid. In a recent review,1 however, we provide many suggestions for how the method could be improved. An acceptable response to our criticisms would be to restrict counterclaims to those supported by reasonable evidence or, even better, to actually test the critical assumptions. For example, one could measure lens density and diffusion at 488 and 527 nm, pupil size, head movements, and so on, on a sample of varying age and quantify the effects of these variables on the Raman signal. If these factors really do have minimal influence as G&B continue to assert, our questions, which are legitimate despite their protestations, could be easily addressed. Providing solid evidence and/or reasoned arguments is undoubtedly a better scientific practice than simply responding negatively and, personally, to being critiqued. Readers can, after all, evaluate the arguments and evidence and come to their own conclusions. What we hope is obvious to all readers is that personal motivations, even if they do exist, are completely irrelevant. For example, G&B suggest that our critique of their method is biased because of a commercial interest by one of the authors. They also describe our review as a “string of shortcomings, misrepresentations, and flawed conclusions.” Such charges do not address content; they simply seek to discredit the source. G&B also suggest that our critique simply duplicated our earlier published correspondence on these issues. It is, however, legitimate to raise similar issues in different papers (G&B, however, often duplicate text and graphs verbatim, which is not an accepted practice). When there was sufficient overlap, we simply referenced earlier sources. Our primary goal with the Raman section of our review was to quantify the confounds that we had identified in earlier papers. G&B also note that our review fails to include any of their rebuttals to our arguments. However, we were careful to note that there is debate regarding these issues and refer the readers to the relevant literature. Most of our criticism of the Raman method is organized around addressing the points that G&B have made in their many publications using the Raman device (often quoting the points G&B made directly). It is, of course, proper scientific practice to cite debate when discussing an issue for which significant debate exists. This is a practice G&B do not follow, e.g., see the most recent papers by Bernstein et al., Ermakov et al., and Ermakov et al.2-4 We have been motivated to write about the Raman method because of the importance of this area and the dramatic claims made by G&B. For example, recently Ermakov et al.3 wrote “The ease and rapidity of Raman MP measurements, the simplicity of the instrumentation, the high accuracy of the measurements, and the lack of significant systematic errors should make this technology attractive for widespread clinical research.” As long as G&B continue to assert validity claims in the absence of evidence, they should expect “incessant” criticism of such claims to continue. This is simply a manifestation of the rigor of our discipline. The validity of a method has nothing to do with its assumed subjectivity or objectivity. One could, for example, argue that the Raman method has an important subjective component because it relies on a subject's own perception of whether they are, or are not, properly aligned. According to Vogt,5 subjective methods are based on the feelings and intuitions of the researcher. Objective methods, in contrast, are based on the “object,” i.e., they are independent of the feelings, beliefs, or desires of the researcher or subject. Based on this definition, all methods currently used to measure MP are objective. Referring to psychophysical methods as subjective seems to reflect a general bias based on the idea that if a test requires a subject's response, it is inherently less accurate despite means for controlling bias, validating the response, and so on. A similar general bias appears to apply in the opposite direction. To some, direct physical methods appear inherently more accurate and less subject to bias despite free parameters, questionable assumptions, a lack of validity testing, and so on. Such biases have no bearing on the actual validity of a method. Validity has to be determined based on a specifiable and reasonable criterion. G&B take some issue with our criterion: matching spectral absorbance profiles with ex vivo curves. This, however, is a common criterion for determining the identity of a signal (i.e., if the spectral curves match, one can be confident that they are measuring carotenoids). George Wald, for instance, originally used this method to identify MP as composed of xanthophylls.6 G&B are, of course, correct in noting that one cannot determine wholly from spectral curves whether they have correctly measured absolute quantity. Essentially, G&B argue that the HFP method yields values that cannot be directly evaluated against known standards and is, therefore, not validated (This argument is, of course, the very one that we have made about the Raman method and has not been refuted by G&B.). When applied to HFP, however, the argument is specious. HFP relies on a derivation strategy rather than a calibration-against-a-standard procedure as is possible in bench scenarios. This derivation is based on a double normalization procedure using wavelength (i.e., relative sensitivity at two wavelengths, those absorbed by MP, e.g., 460 nm, and a reference wavelength that is not, e.g., 560 nm) and retinal site comparisons (two sites, one where MP is present, e.g., the foveal center, and one where MP is known to be optically undetectable, e.g., 7° off the fovea). These normalization procedures are based on the well-established extinction spectrum of MP and the near identity (when using appropriate stimulus conditions) of the relative spectral sensitivity of the response mechanisms at the retinal sites of comparison. The rationale for these assumptions is based on an extensive and well-established literature that is covered in our review. G&B are incorrect in their notion that HFP would be nonlinear as a result of saturation effects. In fact, the inherent perfect linearity of HFP as a measure of MP is an important strength of the method. This is because HFP relies on quantal absorption in outer segments, which are, of course, located behind the MP. Thus, the MP is treated as a filter that merely attenuates the light reaching the receptors. The receptors merely act like photocells responding to quanta. Furthermore, HFP uses a criterion response rather than response amplitude. In other words, the null flicker response is achieved whenever the quantal rate of the variable short-wave component results in a magnitude of response equal to that of the fixed midwave standard. Thus, independent of the amount of MP, the response that determines the match point is of fixed amplitude. The subject merely adjusts the amount of short-wave light until the null point is achieved. A departure from nonlinearity at high OD levels would only result from such extreme light levels as to cause anatomic damage to the eye, an absurdity because such light levels would correspond to an MP OD many orders of magnitude above observed values. As we note within our review, linearity is, of course, a central issue. In comparison even to other physical methods, the Raman method fares particularly poorly in this respect because the signal is returned directly from the molecules of MP. Thus, the deeper layers are screened by the shallower layers. For lower OD levels (approximately <0.30),1 the response could be considered acceptably linear. This range of acceptable linearity could, theoretically, be increased by increasing the intensity of the stimulating laser, but then safety issues would become a serious concern. Using current intensity levels, the response is increasingly nonlinear past 0.30 and this problem cannot be corrected, adjusted, or calibrated away. It is fair to question whether a method designed to measure MP levels returns accurate quantitative estimates for individual subjects. In our review, we discuss this issue with respect to all of the in vivo methods. The Raman method fails seriously even at low levels where the method might be considered acceptably linear. For example, imagine an individual has a true MP OD of 0.30. The Raman signal could be as high as 1500 if none of the signal is lost, or the signal could be near zero if lens absorbance is as high as 0.30 at 488 and 527 nm and scatter is large (such individuals have been identified in the empiric literature, especially at older ages). This permits a range of error that makes interpretation of RCs (Raman count) impossible even on an ordinal scale. In this example, a subject could appear to have either zero MP or average MP density with no way of knowing the true values. G&B note that we did not cite Savage et al.7 and Zagers et al.8 and conclude that “if these more recent results had been used for their calculations, their putative reduction factors would have been much lower in magnitude.” We estimated that the amount of reduction in the Raman signal resulting from lens absorbance between the ages of approximately 20 and 60 was 41%. As we note in our review (footnote h), this estimate is quite similar to that made by G&B,9 which was 38% between the ages of 20 and 60 years. If our “putative reduction factor would have been much lower in magnitude,” it would then be much lower than the estimate made by G&B themselves. Lens absorbance could be easily corrected for if it was simply the average error introduced by the lens that was important. Unfortunately, to interpret individual RCs, knowing the amount of attenuation resulting from absorption by an individual lens is critical. Despite their own estimate of a systematic age-related error of 38%, Ermakov et al.3 wrote that there were no “significant systematic errors” with the method. In another recent paper, Ermakov et al.4 wrote that “To our knowledge there are no serious confounding factors for the technology” These assertions directly contradict G&B's own analysis and conclusions (and inspire our “incessant letters”). The Zagers et al.8 paper was not available when we wrote our review, which was submitted in March 2003. We have since discussed the Zagers et al. and Savage et al. papers in a more recent review1 pointing out that all of the ex vivo data and all of the other in vivo data also support the idea that lens OD at the Raman wavelengths of 488 and 527 nm is significant, increases with age, and is quite variable across subjects at all ages. The fact that two papers conclude that lens absorbance is not as high at the Raman wavelengths as all other papers (>20 other papers using both ex vivo and in vivo methods) find is missing the point. Unless the majority of empiric papers on this topic are incorrect and the lens is perfectly transparent at the relevant wavelengths for all subjects irrespective of age, it will represent an unknown source of error. Although G&B persist in simply claiming that this error is insignificant, that determination cannot be made in the absence of evidence. This is not a “misleading, pejorative statement” that reflects an “unyielding bias.” Based on the author's own conclusions, this is a legitimate question. In fact, confounding by lens absorption alone would invalidate their method, e.g., it could introduce approximately 30% to 40% error of unknown direction. The comparison of HFP and Raman results, cited in the review by Bernstein et al.,2 would have been difficult to evaluate (even if their paper would have been available when writing our review) because virtually no details were provided regarding the comparison. In a more recent review,1 we do discuss the available data regarding the comparison between HFP and the Raman method. The only study that has been published as a full manuscript (including full detail regarding the methods and procedures) is Neelam et al.10 Neelam et al. found that the relation between HFP and RCs was described by an r2 of approximately 0.10. The fact that the methods are measuring the same variable in the same individuals at the same time and only explain approximately 10% of the variance is not a strong argument that the methods agree (particularly if one tried to expand the sample to include the elderly). True agreement would also have to be reflected in the ability to translate the scores such that the relation could be described with an intercept near zero and a slope close to one. It is worth noting also that G&B continue to tout the significance of the correlation between HFP and Raman as evidence for the validity of their method because they argue that HFP is not particularly reliable and may not be valid (see their points 1 and 8). For example, Ermakov et al.3 recently wrote “As a precaution, we avoid media-opacity-related problems by limiting the measurements to subjects with visual acuity better than 20/80. A correlation of our MP Raman responses with data derived by high-performance liquid chromatography and flicker techniques is proof that these precautions are adequate.” We assumed that the authors evaluated the validity of the HFP technique before citing it as evidence. With respect to the former statement, excluding subjects based on visual acuity will not reduce confounds based on individual differences in the lens density of normal subjects because the two have not been shown to be related. It is certainly true, however, that many of the physical methods have shown much more pronounced deflections than is generally seen in the HFP data. How these types of deflections reflect the underlying biology of the retina is an interesting question. It is therefore important to confirm that these topographic irregularities are real. In some cases, for instance, they may reflect artifacts with the method. For example, G&B, when using the Raman method in the imaging mode, report on an individual that they observe had a “central hole in the MP distribution.” Based on what is known regarding the underlying anatomy, it is hard to imagine an actual hole in the center of the MP distribution of a normal healthy individual. Unusual results require careful scrutiny. As we note in a recent review,1 the inherent nonlinearity of the Raman signal could create distortions when used to map MP spatial profiles (as a result of different proportions of the pigment being measured based on location). For example, Hammond et al.13 provide MP spatial density data for a subject whose central MP peak was 1.63 OD (based on an extrapolation from the measured point of 1.35 at a radius of 6 minutes) but whose MP density at 3° was 0.30. Using the Raman calibration curves, the central peak for such a subject would be underestimated by 53%, whereas the 3° site would not be underestimated. In this instance, artifacts with the method would create distortions in the profile that did not, in truth, exist. All areas of science tend to share a focus on methodology. We agree with G&B that the pursuit of multiple methods of measuring MP in vivo is extremely beneficial, including resonance Raman spectroscopy. G&B have claimed that our review of their methodology is biased as a result of the commercial interest of one of the coauthors. One of the coauthors, Dr. Wooten, has built a relatively small number of HFP-based macular densitometers for fellow researchers and colleagues. Our review, however, was in no way designed to promote this instrument nor was the instrument even mentioned. We do not hold a patent or claim ownership of the HFP technique, in specific, or obviously psychophysics, in general. Many laboratories have used HFP to measure MP and results have been published using the method before our laboratory (e.g., Bone).16 Ultimately, critical evaluation must stand on its own merit, irrespective of the author, and is a central feature of our field and an obligation of all scientists. If researchers in the area of macular pigment do not critique their own work, then who will? Billy Wooten, a coauthor of this letter and the original article, is a principal in Macular Metrics Corp. (Providence, RI), which manufactures densitometers. These densitometers use heterochromatic flicker photometry and are used to measure MP. Billy R. Hammond, Jr Vision Science Laboratory University of Georgia Athens, Georgia Billy R. Wooten Bill Smollon Walter S. Hunter Laboratory Brown University Providence, Rhode Island